The multiband cylindrical dielectric resonator antenna with modified microstrip feeding

1 Introduction

Dielectric resonators have gained significant popularity and prominence in the current scenario due to their special characteristics, such as high gain, wide bandwidth, zero conductor losses at higher frequencies, high radiation efficiency, mechanical simplicity, compactness, and so on. Electronic circuits can be easily integrated with DRAs because of their mechanical and thermal stability. These features attract many antenna designers to analyze and study DRAs more extensively. For a long time, circuit applications have been utilizing cylindrical-shaped dielectric resonators. Their high Q-factor and compact size make them ideal for use in filters and oscillators, especially in microstrip technology (Plourdeand and Ren, 1981). S.A. Long conducted the first experimental study of DRA on cylindrical shapes in 1983. The cylindrical shape offers one degree more freedom than the hemispherical shape. With a cylindrical DRA, the designer can choose the most suitable aspect ratio to best realize the desired frequency and bandwidth (Petosa, 2007). Another important advantage of cylindrical DRA is that it supports three operating modes, such as transverse electric (TE), transverse magnetic (TM), and hybrid electric and magnetic (HEM) modes, which permit vast varieties of radiation patterns and provide diversity in terms of practical implementation. An antenna capable of working simultaneously and efficiently at different wireless frequency bands is one of the most important factors in modern communication demands. The efficient rejection of unwanted frequency bands and the yielding of desired frequency bands can only be achieved by a multiband antenna, which is not possible with a single wideband antenna. There has been a huge demand for multiband antennas for the last 2 decades. Plenty of investigations and studies have been conducted on multiband DRAs. Multiband characteristics can be achieved by incorporating stacked DRAs (Sharma and Gangwar, 2015), parasitic-loaded DRAs (Chen et al., 2009; Zaman et al., 2014; Behdad and Sarabandi, 2006), various special types of feeds (Denidni et al., 2005; Hamsakutty et al., 2005; Sharma et al., 2016; Huitema et al., 2011; Pahadsingh and Sahu, 2017), and excitation of higher-order modes (Guha et al., 2014; Chang and Kiang, 2007; Fang et al., 2014; Khan and Ihsan, 2014). But stacked and parasitic-loaded DRAs extend the overall dimension of an antenna. Modifying the geometry and feed shapes can result in increased fabrication complexity. Compared to other techniques, higher-order modes have advantages, but they also increase design complexity. Another popular method for achieving multiband operation is the hybrid cylindrical DRA technique with a dual ‘C'-shaped patch (Sharma and Gangwar, 2016a). However, this design lacks compactness due to its large volume. Two-segment cylindrical DRAs demonstrated in (Raghvendra Chaudhary et al., 2015) along with microstrip feeding result in triple-band operation. But the use of two segments of cylindrical DRA placed one above the other, increases the height of the desired antenna. Sharma and Gangwar (2016b) recommends using three elemental DRAs, metallic patches, and a modified feeding technique for multiband operation. But the usage of three elemental DRs causes it to consume more area. Defected ground structure (DGS) (Azim et al., 2021) with meanderlines (Saeidi et al., 2025), different shapes of slots created in the ground plane (Manish et al., 2022), split ring resonator (SRR), and complementary split ring resonators (Sarfraz et al., 2024) achieve multiband operation. Girjashankar and Upadhyaya (2021) reports that mountable DR can produce multiband operation. But the design complexity and large size make it unsuitable for our modern communication requirements. To overcome the shortcomings of the above-mentioned techniques, this article proposes a suitable technique for operating cylindrical DRA in multiple frequency bands. It provides a compact and multiband CDRA through proper modifications in the geometry and feed. This is achieved through the careful selection of dielectric materials and their corresponding dielectric constants. To enhance the bandwidth of the antenna, an appropriate feed that excites HEM modes is exploited. Multiband is supported by the new geometry and the modified feed.

Section 2 elaborates the design of CDRA and its feed geometry. Section 3 outlines the parametric study conducted to determine the dimensions of the CDRA. Section 4 explains the prototype modeling process and the associated measurements. The designed antenna is compact, low profile, and has very small dimensions of a 10 mm radius and a small thickness of 2.5 mm. It requires a small change in the geometry and an effortless feeding mechanism that doesn't complicate the fabrication process. Also, it provides more than 85% of the radiation efficiency at all the operating frequencies.

2 Antenna design

Figure 1 displays the explicit geometry of the proposed DRA and its top view. The detailed feed geometry and three-dimensional layout of CDRA are presented in Figure 2. As an initial step in the design procedure, a dielectric resonator in cylindrical shape is considered. The cylindrical dielectric resonator is made up of Rogers RT Duroid 6010 material, whose dielectric constant is 10.2. The radius (R) of CDRA is 10 mm, and its height (H) is 2.5 mm. It is firmly mounted on an FR4 substrate material that is 1.6 mm thick and has a dielectric constant of 4.4. The entire setup is placed on a ground plane of length L_D and width W_D. The dimensions of the substrate are the same as that of the ground plane. To make this design compact and low profile, a cylindrical volume of 18 × 6.75 × 2.5 (L × D × H) mm³ has been removed from the center of CDRA. Here ‘L’ and ‘W’ are the length and width of the notch, and ‘H’ is the height of CDRA. This procedure was carried out to decrease the dielectric constant of the entire volume and the Q-factor that is responsible for widening the bandwidth. After a single notch of specific volume, it results in a specific shape that resembles a headphone-shaped CDRA. The significance of its shape is discussed in the next section. The geometry of the flipped ‘F'-type microstripline is elaborated as shown in Figure 2a. It is placed above the substrate and lies below the cylindrical dielectric resonator.

This design utilizes microstripline feeding because of its simple fabrication process and its feed optimization. Also, it gives a wide bandwidth compared to conventional aperture-coupled feeding and doesn’t require drilling, which is mandatory for probe feeding. A microstripline feed is placed at the center of the CDRA with length L₁ and width W₁. In the center feed, two stubs of length L₂ and L₃ and width W₂ and W₃ were placed at the suitable location to enhance the coupling between the feed and the DRA. This flipped ‘F’-type microstripline is used as a frequency tuner to operate the proposed DRA in multiple bands. By adjusting its length and width to a proper dimension, bandwidth can be improved. Microstripline fed at the center of CDRA generates HEM modes. Here HEM stands for Hybrid Electromagnetic modes, which are neither electric nor magnetic. They are purely transverse to the direction of wave propagation. HEM modes are the most commonly used resonant states in CDRAs. Unlike TE or TM modes, HEM modes do not require perfect conducting boundaries to exist and provide excellent radiation into freespace. The resonance frequency of the HEM mode in CDRA is determined by the formula (Petosa, 2007).

f r = 6.324 c / 2 π a 0.27 + 0.36 a / 2 h + 0.02 a / 2 h 2 (1)

Where ‘h’ and ‘a’ denote the height and radius of CDRA. Finally, optimized dimensions were fine-tuned by an exhaustive trial-and-error method using the HFSS 14 software. Table 1 shows the final optimized dimensions of the proposed CDRA.

TABLE 1. Dimensions of the headphone-shaped CDRA.

3 Parametric study

The parametric study is conducted using HFSS 14 simulation software to explore the proper dimensions and characteristics of the intended DRA. Here, two important parametric studies are discussed in detail. The first study is significant in finalizing the structure of the proposed antenna. The return loss characteristics of a simple cylindrical DRA (without a notch) and its resultant DRA (with a notch) are depicted in Figure 3. Initially, a simple CDRA mounted on a flipped ‘F'-type microstripline is simulated. Without notching, CDRA achieves only a single resonance at 6.15 GHz and has a narrow bandwidth of 6.5%. It is not possible to get more than a single band with this conventional CDRA at the desired frequencies. Even though the feed geometry is modified as a flipped ‘F'-type microstripline, conventional CDRA could not achieve multi-bands. But when a cylindrical volume of 18 × 6.75 × 2.5 mm³ (L × D × H) is notched from CDRA, it resonates at three different frequencies, such as 5.915 GHz, 10.76 GHz, and 12.45 GHz, with a large bandwidth of 18.57%, 9.9%, and 15.8%, as given in the figure. Thus, removing a particular cylindrical volume results in multiple band operation, volume reduction, and a headphone-shaped DRA. In addition, this design of CDRA produces return losses as high as −32.7 dB, better impedance matching, and higher gain. The length and width of the above notch are optimized to get the best results and are obtained only when the above-mentioned cylindrical volume at the center is removed. This has been illustrated in Figure 4.

The separating distance or width ‘D’ from the center of DRA can be varied between 5 mm, 6.5 mm, and 8 mm. The desired band of frequencies is not obtained when the distance is less or greater than 6.5 mm, and their bandwidth is comparatively lower. The length of the notch was adjusted to 15 mm, 18 mm, and 19 mm in our investigation.

Figure 5 shows that a wider bandwidth and a good impedance match require a notch length of 18 mm. This is not achieved with other values of length. The next study confirms the structure of the feed geometry. A headphone-shaped DRA located at the center of the microstrip feed with length L₁ and width W₁ was simulated, and the corresponding return loss characteristics are shown in Figure 6. As depicted, center feeding without stub connection doesn’t provide any resonance in the desired frequency bands. When a single stub is connected, it produces dual bands such as 6.1 GHz and 13.1 GHz, with 6% and 5.7%, respectively. As illustrated in the same figure, CDRA associated with two stubs results in good impedance matching and the generation of three different bands with large bandwidth of 18.57%, 9.9%, and 15.8%, respectively. Thus, a flipped ‘F'-type microstripline is employed as a feed for the proposed design. Further parametric studies were conducted for the fine tuning of optimized length and width of stubs.

4 Results and discussion

To verify the design and its theoretical characteristics, headphone shaped CDRA is fabricated and tested. The photograph of the prototype is shown in Figure 7. The fabricated antenna was measured using a PNA E8362B network analyzer.

Figure 8 shows the graph of the comparison between simulated and measured return loss characteristics. They are in very close agreement. Table 2 shows the measured results of headphone-shaped CDRA. Fabricated DRA resonates at three different frequencies, such as 5.915 GHz, 10.76 GHz, and 12.45 GHz, and is suitable for WLAN frequency bands and other frequency bands like the ‘X’ band and the ‘Ku’ band. It yields a bandwidth of 18.57%, 9.9%, and 15.8% in the lower, middle, and upper frequencies, respectively. It is observed that HEM modes are excited at resonant frequencies 5.915 GHz, 10.76 GHz, and 12.45 GHz. HEM modes are denoted as HEM _mn**d**, where ‘m’ represents the azimuthal variation, ‘n’ defines the radial variation, and ‘d’ indicates the half cycle variation along the cylinder axis. Coupling between the central feeding of microstripline and cylindrical DRA induces HEM_11**δ**, HEM_12**d**, and HEM_13**d** modes. Here HEM_11**δ** acts a like horizontal magnetic dipole producing broadside radiation with a wide beamwidth. HEM_12**d**, and HEM_13**d** are higher order modes that provide higher directive gain. The proposed CDRA was fabricated using a Rogers RT Duroid 6010 dielectric material slab.

TABLE 2. Measured results of headphone-shaped CDRA.

Therefore, DRA was made with a thickness of 2.5 mm without the need to stack layers one above the other. After accurate machining of the headphone shape, the dielectric material bonds with the substrate material in the correct position using the commercially available Araldite glue. The glue was chosen with a very low dielectric constant to avoid deviation from the results.

The E-plane and H-plane radiation patterns of the proposed DRA were measured in an anechoic chamber. Radiation patterns of simulated and measured results of headphone-shaped CDRA are shown in Figures 9, 10.

As demonstrated, the proposed CDRA has linear polarization and radiates in a broad direction. Throughout the entire frequency band of operation, the proposed antenna provides stable radiation. It is absolutely true that the amplitude level of co-polarization is higher than that of cross-polarization in both E-plane and H-plane radiation patterns. Furthermore, a study was conducted to examine the gain characteristics of headphone-shaped CDRA. Peak gain is calculated along the direction of maximum radiation. A horn antenna is used as a reference antenna for gain measurement in an isolated anechoic chamber. Figure 11 shows the simulated and measured peak gain of the suggested dielectric resonator antenna. Fair agreement is reflected between the simulated and measured results. Measured gain ranges from 4.5 dBi to 7.05 dBi over the operating frequencies. The proposed CDRA achieves a measured peak gain of 7.05 dBi at 10.88 GHz. Figure 12 shows the plot of the measured and simulated radiation efficiency of CDRA. To obtain radiation efficiency, the Wheeler method is used, which compares the reflection coefficients of an antenna measured in free space to those measured in the chamber. It is the easiest method and does not require a reference antenna.

It is obvious that more than 87.5% of theoretical radiation efficiency is achieved, and more than 85% of radiation efficiency is achieved practically over the operating range of frequencies. Table 3 analyzes the comparison between the proposed CDRA and other recently published DRAs given in the references. Even though (Sharma and Gangwar, 2016c) has a broad bandwidth in the lower and middle bands, it consumes 8 mm of thickness, which is too large and cannot be suitable for compact wireless applications. Despite this, the proposed DRA maintains a thickness of only 2.5 mm while achieving a wider bandwidth.

TABLE 3. Performance comparison of proposed antenna and other published antennas.

5 Conclusion

Thus, a novel, shaped cylindrical dielectric resonator antenna was thoroughly investigated, measured, and implemented for multiband operation. The flipped ‘F'-type microstripline and the new shape of geometry support triple band operation with ample bandwidth in all three bands. Good agreement was achieved between the simulated and measured results. The proposed design has an extremely thin profile, is compact, and achieves good gain and radiation efficiency throughout the working bands. Simple geometry and an easy feed mechanism are the key features of the proposed design. It is undeniably the optimal choice for WLAN, ISM, ‘X’ and ‘Ku’ bands, as well as satellite applications.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Francesca Benassi, University of Bologna, Italy

Reviewed by: Rasmita Sahu, Veer Surendra Sai University of Technology, India

Yassine El Hasnaoui, Universite Ibn Tofail Ecole Nationale des Sciences Appliquees de Kenitra, Morocco